1. Field of the Invention
The present invention relates to a chain code generating apparatus and a method thereof, more specifically to a chain code generating apparatus and a method thereof that can generate a chain code in real time by performing an outline search of an analysis object in an input image.
2. Background Art
In some industrial fields, their own functions can be performed enough by using only binary images having two pixel values instead of color images or gray scale images which require a lot of memories and large bandwidths. For example, their own processes can be executed in the fields of character recognition, photographed images of a printed circuit board and medical images such as X-ray. In these fields, using the binary image having two pixel values makes it possible to analyze interest information in order to perform desired functions such as character recognition, error-checking of a printed circuit board and part-recognition of factorial assembling line.
Such a binary image system can use fewer memories than color images or gray scale images and employ intuitive algorithms. Also, the binary image system can be processed by using easy logic operations instead of complex logic operation, which results in the quick operation speed. This is because the binary image system has been still popular in the general industrial fields although today's computing system is dramatically developed.
In particular, the binary image system is more useful in the system environments which can be controlled to allow an object and surroundings to be easily separated. The environment separation can be easily performed by allowing the number of objects of an image to be limited and adequately controlling lightings of the system environment. For example, since the lighting control and the number adjustment of the objects included in the image are very simple things, the binary image system is very popular in the actual industrial spots such as factories.
In the binary image, the most important items of information are the shape, position and direction of an object portion (i.e. a portion occupied by an object in an image). These items of information can be extracted by using outline information related to a part in which a pertinent object is in contact with its surrounding. Accordingly, it is unnecessary to use overall image information in order to analyze an analysis object in a binary image, but using the outline information can perform the analysis of the analysis object efficiently.
In other words, the processed result can be acquired more quickly although fewer system resources such as memories and CPUs are used. Accordingly, the application field based on the binary image does not use the binary image itself as it is for the image analysis but convert the form of outline information of the analysis object in the binary image to a suitable form to use the binary image.
A chain code method is one of the well-known methods for representing outlines. A chain code codes the relation between each pixel forming the outlines of an analysis object in an image to represent the analysis object. Here, the relation between each pixel is referred to as the position correlation between a current pixel and a next-searched outside pixel, which form the outline.
By using the chain code, it is possible to recognize the position correlation between each of the pixels which start from a start point pixel of the analysis object and arrive at the start point pixel according to the outline of the analysis object. Inversely, since the chain code has information related to the position correlation between each pixel included in the outline of the analysis object as described above, the origin binary image can be recovered from the chain code.
In the conventional art, one image processing system generates a chain code and analyzes the generated chain code. In other words, even through generating the chain code and analyzing the generated chain code does not belong to the main work of the image processing system that performs the general imaging processing of an input image but merely belong to an additional work of the image processing system, it consumes too more system resources than needs to allow the single image processing system to process the two operations.
This requires a chain code generating apparatus, which can reduce the processing road of an image processing system from the viewpoint of a general system by allowing a separately mounted compact hardware device to execute the processes of converting an inputted original image to a binary image and acquiring a chain code from the binary image.
As such, if the separately mounted compact hardware device is allowed to quickly generate a chain code, the image processing system can concentrate its system resources on an upper algorithm and complex image processing operations by using the chain code. Accordingly, the system can more efficiently perform the processing operations.